Application of remote sensing data to measure erosion on rehabilitated landforms at the Abydos mine

Authors: Crisp, H; Mackenzie, S; Gregory, S; Sprenkels, T; Slabber, A Download Paper Open access courtesy of:

DOI https://doi.org/10.36487/ACG_repo/2415_72

Cite As:
Crisp, H, Mackenzie, S, Gregory, S, Sprenkels, T & Slabber, A 2024, 'Application of remote sensing data to measure erosion on rehabilitated landforms at the Abydos mine ', in AB Fourie, M Tibbett & G Boggs (eds), Mine Closure 2024: Proceedings of the 17th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 1005-1018, https://doi.org/10.36487/ACG_repo/2415_72



Abstract:
Monitoring stability and erosion is a critical requirement for rehabilitated mining landforms. Monitoring efforts have shifted from transect based and qualitive assessments, to remotely sensed landform-scale unmanned aerial vehicle (UAV) surveys. Landform-scale UAV surveys enable spatial inventories of erosion features and their geometric parameters to be developed and monitored overtime. UAV captured data can also be used to determine the causes and severity of erosion, to inform remedial action plans and aid the development of completion criteria. Capturing high-resolution elevation data is the key for conducting landform-scale UAV erosion monitoring. There are two main approaches for capturing and generating high-resolution elevation models: photogrammetry and light detection and ranging (LiDAR). LiDAR data is more suitable for vegetated areas as it can provide ground detail below plant canopies. As vegetation grows, photogrammetrically derived elevation models that include vegetation obscure erosion features and prevent interpolation of feature parameters, particularly gully volumes. One challenge with using LiDAR data relates to how accurately the data is classified. In our case study a semiautomated iterative scale and relative height filtering approach was applied to accurately classify LiDAR data into ground and non-ground classifications, to generate a bare earth digital elevation model. Through a combination of slope visualisation and manual assessment, erosion features were then digitised and their geometries, including gully volumes, were calculated. Erosion features were categorised to differentiate between minor rills and substantial erosion gullies that pose a risk to landform stability. In our case study, we analysed annual landform-scale UAV survey data from Atlas Iron’s Abydos mine in the Pilbara region of Western Australia. The high-resolution elevation data and spatial inventories of erosion features collected from the rehabilitated landforms has allowed us to measure and track temporal changes in gully parameters (gully length, area, and volume) and to understand change in overall landform stability. This insight has enabled causes of erosion gullies to be identified, targeted remedial action plans to be developed and meaningful completion criteria to be refined; all of which have contributed to positive longterm stability outcomes.

Keywords: remote sensing, erosion, LiDAR, unmanned aerial vehicle (UAV), remediation, waste rock landforms (WRL)

References: